RESEARCH ARTICLE◥

NEUROSCIENCE

Lattice system of functionally distinctcell types in the neocortexHisato Maruoka,* Nao Nakagawa,* Shun Tsuruno,* Seiichiro Sakai,*Taisuke Yoneda,* Toshihiko Hosoya†

The mammalian neocortex contains many cell types, but whether they organizeinto repeated structures has been unclear. We discovered that major cell typesin neocortical layer 5 form a lattice structure in many brain areas. Large-scalethree-dimensional imaging revealed that distinct types of excitatory and inhibitoryneurons form cell type–specific radial clusters termed microcolumns. Thousands ofmicrocolumns, in turn, are patterned into a hexagonal mosaic tessellating diverseregions of the neocortex. Microcolumn neurons demonstrate synchronized in vivoactivity and visual responses with similar orientation preference and oculardominance. In early postnatal development, microcolumns are coupled by celltype–specific gap junctions and later serve as hubs for convergent synaptic inputs.Thus, layer 5 neurons organize into a brainwide modular system, providing a templatefor cortical processing.

The mammalian neocortex is densely popu-latedbydiverse types of excitatory and inhib-itory neurons, each with specific molecularand cellular properties, synaptic connec-tions, and in vivo functions. Whether neo-

cortical cell types organize into repeated structuresrepresenting common motifs for informationprocessing has been poorly understood. Corticalcolumns, including orientation columns in the vi-sual cortex and barrels in the somatosensory cor-tex, have patterned structures, but their cellularand circuit-level organization is largely unsolved(1). Moreover, cortical columns are restricted tospecific cortical areas and therefore do not rep-resent brainwide structural motifs.In neocortical layer 5, subcerebral projection

neurons (SCPNs) are one of the two major ex-citatory neuron types and have well-definedanatomical and genetic specifications. SCPNsconstitute the major cortical output pathway,sending massive axonal projections to subcorticaltargets including the pons, spinal cord, and su-perior colliculus (2). Prior studies described a lo-cal arrangement of SCPNs in radial clusters witha diameter of one to two cells and tangentialdistances of a few cell diameters (3, 4). Theseclusters, here termed microcolumns, have beenreported in visual and somatosensory corticalareas in mice (3) and in language areas in hu-mans (4). To study the cellular organization ofneocortical layer 5, we conducted structural andfunctional analyses of SCPN microcolumns andinvestigated whether other cell types organizeinto microcolumnar structures.

Spatial organization of major cell typesin layer 5We examined the three-dimensional organiza-tion of SCPN microcolumns in the mouse brain.SCPNswere retrogradely labeled by injecting fluo-rescent tracers into the pons (Fig. 1A and fig. S1A).After fixation and clearing, the neocortex wasscanned using two-photon microscopy (Fig. 1B).Confirming previous results (3), we observed SCPNmicrocolumns in visual and somatosensory corti-ces (Fig. 1, C andD). The radial alignment of SCPNswas also conserved in the motor cortex (Fig. 1E).Statistical analyses showed a microcolumnar or-ganization in nearly all examined neocortical re-gions (Fig. 1F and fig. S1, B to L). The orientationof microcolumns gradually changed along thecortex (Fig. 1F and fig. S1M) but remained ap-proximately parallel to apical dendrites (fig. S1,N to P, andmaterials andmethods). The radiusof microcolumns was ~10 mm in visual, somato-sensory, and motor cortices (Fig. 1G).Analyses of the organization of microcolumns

have been performed previously in brain slices(3, 4), but their two-dimensional lateral arrange-ment in the cortex has not been determined. Atwo-dimensional Fourier analysis of the SCPNdistribution revealed a periodicity of 30 to 45 mm(P < 0.001 in three of three mice; fig. S1, Q to S).We further analyzed the periodicity with a cor-rection formicrocolumn tilt (fig. S1T). Tangentialsection images (Fig. 1, H and I) and the autocor-relogram of the SCPN distribution (Fig. 1J) sug-gested an approximately hexagonal pattern, whichwas observed in multiple cortical areas (Fig. 1K).We found a sixfold symmetry (P < 0.01; fig. S1, Uto X) but no other rotational symmetries. Consist-ently, the two-dimensional power spectrum hadsix peaks ranging from 24 to 30 cycles/mm, twoof which were located on the anterior-posterior

axis and the other four of which were at lateralpositions (Fig. 1L; computed for areas containing≥2,000microcolumns). Individualmicrocolumnswere positioned near the intersections of thethree waves reconstructed from the six peaks ofthe power spectrum (Fig. 1, H and I).Layer 5 contains another major type of excit-

atory neuron that innervates the cerebral cortex:cortical projection neurons (CPNs) (2).We labeledCPNs in Tlx3 (T-cell leukemia homeobox 3)–cre/Ai9 mice (5) (green in Fig. 2A and fig. S2, A to F)and, in parallel, visualized SCPN microcolumnsby retrograde labeling (magenta in Fig. 2, A andB, left). In layer 5b—the lower part of layer 5,where SCPNs are present—the density of CPNsradially aligned to SCPNs was lower than theaveragedensity of CPNs (P<0.01; Fig. 2B,middle),indicating that CPNs were excluded from SCPNmicrocolumns. Moreover, CPNs were radiallyaligned to each other in an orientation parallelto SCPN microcolumns (P < 0.01; Fig. 2B, right).Thus, CPNs are organized into cell type–specificmicrocolumns that interdigitatewith SCPNmicro-columns. CPNs in layer 5a also adopted a micro-columnar arrangement (fig. S2G).We also investigated the arrangement of the

twomost prevalent inhibitory neuron types in layer5—parvalbumin-expressing (PV+) and somatostatin-expressing (SOM+) cells (6)—using fluorescent im-munostaining in three-dimensional samples (7 )(Fig. 2, C to F). PV+ and SOM+ cells aligned radi-ally with SCPNs (P < 0.01; Fig. 2, C and D) butnot with CPNs (Fig. 2, E and F, and fig. S2H),indicating a selective alignment of inhibitory neu-rons to excitatory neuron microcolumns.

In vivo neuronal activityof microcolumns

In vivo microcolumn activity has been only in-directly inferred from studies of immediate earlygene expression in fixed slices (3). We thereforeinvestigatedmicrocolumn activity in vivo using anadeno-associated viral (AAV) vector that expressesthe Ca2+ indicator G-CaMP6 (8) almost exclusivelyin SCPNs (Fig. 3, A and B, and fig. S3), most likelyowing to tropism. Ca2+ signalswere obtained fromawake mice using two-photon volume imaging(120 to 240 mm thick, 1.6 to 2.4 volumes/s). Theorientation of SCPN microcolumns was approxi-mated from the axes of the apical dendrites. Datafrom four representative SCPNs in the binocularvisual cortex, recordedwithout visual stimulation,are shown in Figure 3, C to G. Cells 1 to 3 (tan-gential distance of <15 mm) showed synchronousCa2+ signals, whereas cell 4 (tangential distanceto the other three cells of >25 mm) exhibited nosynchronization with other cells (Fig. 3E). In ac-cord, the temporal correlation of Ca2+ traces washigher among cells 1 to 3 than between cell 4 andother cells (Fig. 3G). We calculated the averagecorrelation as a function of the tangential distance(Fig. 3H, left). The actual correlation values attangential distances of <15 mmwere greater thanthose for random surrogates (Fig. 3H), whereasthe actual values at tangential distances of >20 mmwere almost at the level of those for the surro-gates (Fig. 3H), indicating synchronized activity

RESEARCH

Maruoka et al., Science 358, 610–615 (2017) 3 November 2017 1 of 6

RIKEN Brain Science Institute, 2-1 Hirosawa, Wako, Saitama351-0198, Japan.*These authors contributed equally to this work. †Correspondingauthor. Email: hosoya@brain.riken.jp

on March 17, 2020

http://science.sciencem

ag.org/D

ownloaded from

within radially aligned SCPNs (Fig. 3, H and I, andfig. S4, A to D). We also confirmed significantsynchronization within individual microcolumns(materials andmethods). The observed correlationwas not caused by light contamination (Fig. 3, Eand F; fig. S4, E to G; andmaterials andmethods).Similar results were obtained in the primary so-matosensory and motor cortices (Fig. 3, J and K;fig. S4, A to G; and materials and methods).Although the superficial layers of the mouse

visual cortex show weak clustering of neuronswith similar response properties (9, 10), cellu-lar clustering for visual functions in layer 5 ispoorly understood. We analyzed the orienta-tion preference and ocular dominance of SCPNmicrocolumns in the binocular visual cortex.

We presented drifting gratings with six differ-ent orientations (Fig. 4A) and determined thepreferred orientation for each SCPN that exhib-ited orientation-selective responses. The differ-ence in the preferred orientation of SCPN pairswith a tangential distance of >20 mmwas similarto that of randomly selected pairs (~45°; Fig. 4C,left). In contrast, SCPN pairs with a tangentialdistance of <10 mm and pairs within individualmicrocolumns had a significantly smaller differ-ence (~32°; Fig. 4C, left; fig. S4H; and materialsand methods). The similarity was observed evenwhen the radial distance was as large as ~80 mm(Fig. 4C, right, and fig. S4H).We also determinedthe ocular dominance index (ODI) by stimulat-ing both eyes alternately (Fig. 4B). The ODI was

similar for SCPN pairs with a tangential dis-tance of <10 mm and within individual micro-columns (Fig. 4D and materials and methods).

Chemical and electrical synapticconnections of microcolumns

We investigated the synaptic circuits that could co-ordinate microcolumnar neuronal activity. Whole-cell patch clamp recordings were obtained fromtwo to four enhanced green fluorescent protein(EGFP)–labeled SCPNs in acute slices preparedfrom the visual cortex of Crym-egfpmice (3) at ~4postnatal weeks (Fig. 5A), when in vivo synchron-ized activity was already present (fig. S5A). Wefirst examined mutual connections and failed todetect preferential connections between radially

Maruoka et al., Science 358, 610–615 (2017) 3 November 2017 2 of 6

Fig. 1. Lattice orga-nization of SCPNmicrocolumns. A,anterior; P, posterior;M, medial; L, lateral.(A) SCPNs werelabeled withretrograde tracers. L5,layer 5; Sup, superiorcolliculus. (B) Labeledbrain at age 5 weeks.Gamma correctionwas applied uniformly.(C to E) Sagittal [(C)and (D)] and coronal(E) optical sections of(B). (F) Detection ofmicrocolumnar align-ment. (Top left) Thedensity of SCPNs(circles) in the twocylindrical volumesrelative to otherSCPNs wasdetermined forvarious orientations.(Right) Each panelshows the result forcells around thecorresponding squarein the bottom leftpanel. Minimum-maximum ranges aregiven in the materialsand methods. (Bottomleft) Estimatedmicrocolumn orienta-tions [data from thebrain in (B)].(G) Structure ofmicrocolumns. Each cortical area was divided into multiple subregions ofabout equal size. In each subregion, the SCPN density was measured atdifferent tangential distances from other SCPNs (top left) and normalizedto the average density in the subregion. Green line and shading, meanand SEM, respectively, across subregions (two mice at 5 weeks of age).The troughs at 16 to 24 mm were significant (P < 0.005 for the three areas,two-tailed sign test against 1). (H) Tangential section image of the brain in(B). Retrogradely labeled SCPNs are shown in black. Thickness, 200 mm.(I) Enlarged image. Red dots are the estimated centers of microcolumns.

Gamma correction was applied uniformly in (H) and (I). (J) Autocorrelo-gram of the SCPN density distribution in (H) (1539 cells). (K) Autocor-relograms for the visual (2676 cells) and somatosensory (3982 cells) areasof the brain in (B). (L) (Left) Two-dimensional power spectrum of thedensity distribution of 15,765 SCPNs in the brain in (B). Colored circlesrepresent the centers of the peak frequency components used toreconstruct the waves in (H) and (I). (Right) Power spectrum for 17,307SCPNs in another mouse. Colored circles are at the same positions asthose in the left panel.

RESEARCH | RESEARCH ARTICLEon M

arch 17, 2020

http://science.sciencemag.org/

Dow

nloaded from

Maruoka et al., Science 358, 610–615 (2017) 3 November 2017 3 of 6

Fig. 2. Cell type–specific microcolumnarorganization of CPNs and inhibitory neurons.Analyses of layer 5b cells in mice at age 5 to12 weeks. Photographs show the somatosensoryarea. White dashed lines mark the borderbetween layers 5a and 5b. (A and B) Analysis ofSCPNs and CPNs. Colored lines in (B) show celldensity for 31,304 SCPNs and 12,560 CPNs intwo mice (determined similarly to the resultsshown in Fig. 1G). (Left) The mean and SEM(shading) calculated among multiple subregionsin the cortex. The trough at 15 to 25 mm wassignificant (P < 3.1 × 10−5, two-tailed sign testagainst 1). (Middle and right) Gray, 100-surrogatedata generated by random positioning of CPNs.Dashed lines, highest and lowest 2.5% of surro-gates. Black line, median of surrogates. Thetrough in the right panel at 15 to 20 mm wassignificant (P < 0.01). (C to F) Analysis ofinhibitory neurons, shown similarly to the middlepanel in (B) (data from three mice): (C) 5727SCPNs and 1990 PV+ cells, (D) 6179 SCPNs and1793 SOM+ cells, (E) 7601 CPNs and 2830 PV+

cells, and (F) 8772 CPNs and 3249 SOM+ cells. In(E) and (F), the orientation of microcolumns wasestimated from that of the apical dendrites.

Fig. 3. Synchro-nized activity inSCPN microcol-umns. Data frommice at age 11 weeks.(A) Diagram ofimaging. Data fromthe binocular visualcortex are shown in(B) to (I). (B)Expression ofG‐CaMP6 in SCPNs.(C) Example cells(1 to 4). (D) Tempo-rally averagedimages of the cellsin (C), top view.Green dotted lines,cell contours.(E) Ca2+ traces ofthe cells in (C).Magenta lines andarrows, synchro-nized peaks of cells1 to 3. Cyan lines,large peaks in onlyone of cells 1 to 3,indicating that lightcontamination wasundetectable. DF/F,relative change influorescence. (F) Dataat “F” in (E). Left, Ca2+ traces. Right, fluorescence images (top view) atthe time frames indicated by dotted lines in the left panel. Green dottedlines, cell contours. Peaks in Ca2+ traces were accompanied by a signalincrease across the entire cell body, indicating that they representneuronal activity but not light contamination. (G) Correlation coefficientsof Ca2+ traces for cells 1 to 4. (H and I) Dependence of the average

correlation on the distance (1209 cells in three mice). (H) Dependence ontangential distance. Green, actual data. Gray, data from 1000 randomsurrogates. Black dotted and solid lines, top 5% and median of randomsurrogates, respectively. (I) Dependence on radial distance. Pairs witha tangential distance of <10 mm were analyzed. (J) Somatosensory cortex(1529 cells, three mice). (K) Motor cortex (1302 cells, three mice).

RESEARCH | RESEARCH ARTICLEon M

arch 17, 2020

http://science.sciencemag.org/

Dow

nloaded from

aligned SCPNs (fig. S5B), consistent with previ-ous findings (11, 12). We next examined commonsynaptic inputs to a pair of neurons, which gen-erate synchronized excitatory postsynaptic cur-rents (EPSCs) with a typical time difference of<1ms (13–15) (Fig. 5, B andC, and fig. S5, C andD).When all EPSCs were included for the analysis ofsynchronized EPSCs, we found no preference forradially aligned SCPNs (fig. S5E).We subsequentlyexamined large EPSCs, which are particularly im-portant for mammalian brain function (16–18).When we analyzed the largest 7.4% of EPSCs in-

duced by presynaptic spikes (corresponding tothe largest 2% of all recorded EPSCs, includingthose not induced by presynaptic spikes; Fig. 5D;fig. S5, F and G; andmaterials andmethods), theprobability that radially aligned SCPNs (tangen-tial distance of <7.5 mm) had synchronized syn-aptic activity was higher than that expected foruniform random connections (P = 0.0036, one-tailed binomial test; Fig. 5E), andmore than seventimes that for pairs with only slightly longer tan-gential distances (7.5 to 22.5 mm; P = 0.0012, one-tailed Fisher’s exact test; Fig. 5E). The results were

robust against changes in the parameters used forthe analyses (fig. S5H). In contrast, we found nopreference for tangentially aligned SCPNs (P =0.63, one-tailed binomial test, and P = 0.51, one-tailed Fisher’s exact test; Fig. 5F). These resultssuggest that SCPNs in individual microcolumnspreferentially receive strong synaptic inputs fromcommon presynaptic neurons (materials andmethods).Microcolumns are present at postnatal day 6

(P6) (3), when chemical synapses are still infre-quent (19, 20), suggesting the possibility thatmicrocolumn neurons have cellular interactionsother than chemical synaptic connections duringcortical development. We therefore investigatedgap junction–mediated electrical coupling, whichis implicated in the development of neuronal cir-cuits (21), in acute visual cortex slices (Fig. 6, Aand B, and fig. S6, A to D). At P6 to P7, abouthalf of neighboring SCPN-SCPN and CPN-CPNpairs exhibited electrical coupling (37 of 75 SCPN-SCPN pairs, 49%; 44 of 80 CPN-CPN pairs, 55%;Fig. 6C), whereas only 12% of neighboring SCPN-CPN pairs were coupled (7 of 58 pairs; Fig. 6C).The coupling probabilities of SCPN-SCPN andCPN-CPN pairs were significantly higher thanthat of SCPN-CPN pairs (P < 10−5 for both, two-tailed Fisher’s exact tests). Further, the couplingcoefficients of SCPN-SCPN and CPN-CPN pairswere significantly greater than those of SCPN-CPN pairs (Fig. 6C). Electrical coupling becameundetectable by the end of the second postnatalweek (Fig. 6C).In the radial orientation, the coupling coeffi-

cient of SCPN-SCPN pairs was largely indepen-dent of distance up to 50 mm (Fig. 6, D and E). Incontrast, in the tangential orientation, the cou-pling coefficient rapidly decreased and approachedzero at a distance of >30 mm (Fig. 6, D and E).Consequently, radially aligned pairs had largercoupling coefficients than tangentially alignedpairs at the same distance (Fig. 6E). Moreover,pairs with a tangential distance of <15 mm hadgreater coupling coefficients and coupling prob-abilities than those with larger tangential dis-tances (Fig. 6F). A similar radial bias was foundfor CPN-CPN coupling (fig. S6, E and F). These re-sults suggest that gap junctions preferentiallycouple neurons within individual microcolumns(materials and methods).In the developing neocortex, clonally related

excitatory neurons are preferentially coupled bygap junctions, but the coupling becomes in-frequent by the end of the first postnatal week(~2% at P6) (21). In contrast, radially alignedSCPNs are mostly nonsisters (3) and frequentlycoupled at P6 to P7 (Fig. 6, D to F), suggestingthat the coupling observed in this study occurredbetweennonsister pairs. In accord, half of neigh-boring nonsister SCPN pairs had electrical cou-pling at P6 to P7 (18 of 38 pairs; fig. S6, G to J).

Discussion

We discovered that wide areas of neocorticallayer 5 are organized into a cellular lattice systemcomposed of cell type–specific microcolumns(fig. S7). The functionalmodularity suggests that

Maruoka et al., Science 358, 610–615 (2017) 3 November 2017 4 of 6

Fig. 4. Orientation preference and ocular dominance of SCPN microcolumns. Data from thebinocular visual area at age 10 to 11 weeks. (A and B) Example of radially aligned SCPNs. (A) (Left)G-CaMP6 labeling. (Right) Individual (light blue) and average (dark blue) responses to grating stimulidelivered to the contralateral eye. (B) Responses to contralateral and ipsilateral stimulation,averaged across orientations. (C) Difference in the preferred orientation of SCPN pairs (1621orientation-selective SCPNs in six mice). (Left) Median values among cell pairs plotted against thetangential distance. Red, actual data. Gray, data from 1000 random surrogates. Black solid anddotted lines, median and bottom 5% of random surrogates, respectively. P = 0.002 for the tangentialdistance of <10 mm. (Right) Dependence on the radial distance. Pairs with a tangential distance of<10 mm were analyzed. The radial bin width was 50 mm. (D) Mean difference in the ocular dominanceindex (ODI), shown similarly to (C), where the blue line is actual data (2136 visually responsiveSCPNs in seven mice). P = 0.007 for the tangential distance of <10 mm in the left panel.

RESEARCH | RESEARCH ARTICLEon M

arch 17, 2020

http://science.sciencemag.org/

Dow

nloaded from

single microcolumns perform elementary cir-cuit functions that collectively constitute large-scale parallel processing. The descriptions ofmicrocolumns in multiple cortical areas in mice[(3) and this study] and humans (4) suggest thatthe lattice system is a neuronal architecture com-mon to cortical functions as diverse as sensory,motor, and language processing. The coordi-nated in vivo activity of SCPN microcolumnsand their convergent inputs indicate that theyconstitute a brainwide system of modular, re-

peated synaptic circuits and discrete corticaloutput channels.Several mammalian species, but not rodents,

possess ocular dominance columns and orien-tation columns. The typical width of an oculardominance column is ~500 mm, and the pre-ferred orientation of orientation columns changesgradually across the cortical surface. In mice,neighboring microcolumns (distance of ~40 mm)had no apparent similarity in visual responses.We hypothesize that in species that have orien-

tation columns and ocular dominance columns,neighboring microcolumns may be progressivelyarranged to have similar functions, thereby con-tributing to the anatomy of known cortical col-umns. Orientation columns have a width roughlysimilar to microcolumn spacing (1) and exhibita hexagonal arrangement (22); therefore, theymay be constructed on the basis of the latticesystem.Previous studies demonstrated that clonally

related neurons show electrical coupling in the

Maruoka et al., Science 358, 610–615 (2017) 3 November 2017 5 of 6

Fig. 5. Convergent strong inputs to SCPN microcolumns.(A) A visual cortex slice of a P24 Crym-egfp mouse. SCPNs andrecorded cells were labeled by EGFP expression and biocytininjection, respectively. White arrows indicate recorded neurons.(B) Current traces (top) and EPSC onsets (bottom) of a SCPN pair.Asterisk, coincident EPSCs. (C) Cross-correlogram of EPSC ratesof the pair in (B). Bin size, 2 ms. (Inset) Cross-correlogram shown ona longer time scale. (D) Tangential and radial distance betweensimultaneously recorded SCPN pairs. Red and gray dots indicateSCPN pairs with and without synchronized EPSCs, respectively(n = 183 pairs; P21 to P28). (E) Black, probability that a SCPN pairhad synchronized EPSCs. Error bars, 95% confidence intervalsestimated using the binomial distribution. Blue, average probability.Fractions indicate the number of pairs with synchronized EPSCsout of the total number of recorded pairs. Black numbers arefor pairs in the first (<7.5 mm) and second (7.5 to 22.5 mm)bins. Blue numbers are for all pairs. (F) The same analysisas in (E) but for the radial distance. **P < 0.01; n.s., notsignificant (P ≥ 0.05); one-tailed Fisher’s exact tests.

Fig. 6. Cell type–specific microcolumnarelectrical coupling during development. (A andB) Recordings in visual cortex slices of Crym-egfpmice at P6 to P7. Numbers indicate individualneurons. Green, EGFP-expressing SCPNs. Magenta,retrogradely labeled CPNs. Light blue and white,recorded neurons labeled by biocytin injection.(A) A slice at P6. (B) (Left) Recorded neurons.(Right) Black, hyperpolarizing pulses injected intoa neuron. Dark blue, the average membranepotential response of the other neuron. (C) Couplingcoefficients of neighboring pairs. Center-to-centerdistances, <25 mm (P6 to P7) and <30 mm (P10 toP11 and P14 to P15). Numbers of recorded pairsare shown on the horizontal axis. The vertical axisis truncated. (D) Distribution of coupled (filledcircles) and noncoupled (open circles) SCPN-SCPN pairs (n = 123 pairs). (E) Dependence ofcoupling coefficient on the distance. Triangles,average coupling coefficient of pairs with atangential distance of <20 mm, plotted against theradial distance. Circles, average couplingcoefficient of pairs with a radial distance of<20 mm, plotted against the tangential distance.Error bars, SEM. In the statistical tests betweenorientations, pairs grouped to both orientations wereexcluded. (F) Comparisons between pairs with adifferent tangential distance. Fractions indicate thenumbers of coupled pairs out of all tested pairs.Error bars, 95% confidence intervals for thebinomial distribution. Right panel in (F), two-tailedFisher’s exact test; other panels, two-tailed Mann-Whitney-Wilcoxon tests. ***P < 0.001; **P < 0.01;*P < 0.05; n.s., not significant (P ≥ 0.05).

RESEARCH | RESEARCH ARTICLEon M

arch 17, 2020

http://science.sciencemag.org/

Dow

nloaded from

early neonatal stage (21) and later exhibit similarorientation preference (23, 24). In contrast, ourfindings show that microcolumns are composedof clonally unrelated cells (3) that have specificelectrical coupling during P6 to P7, when corticalsynapses are being generated (19, 20). The tran-sient couplingmay synchronize neuronal activityand promote the development of microcolumn-specific circuits throughHebbian-likemechanisms.Gap junctions may also facilitate microcolumnarclustering through cell adhesion properties (25).Microcolumns might be structurally linked to“neuronal domains”—radial neuronal clusters ob-served in neonatal cortex that are suggested tohave gap junction coupling (26–28). Because neu-ronal domains spanmultiple cortical layers, micro-columns might also be present in other corticallayers.

REFERENCES AND NOTES

1. J. C. Horton, D. L. Adams, Philos. Trans. R. Soc. Lond. B Biol.Sci. 360, 837–862 (2005).

2. B. J. Molyneaux, P. Arlotta, J. R. Menezes, J. D. Macklis,Nat. Rev. Neurosci. 8, 427–437 (2007).

3. H. Maruoka, K. Kubota, R. Kurokawa, S. Tsuruno, T. Hosoya,J. Neurosci. 31, 18522–18542 (2011).

4. K. Y. Kwan et al., Cell 149, 899–911 (2012).5. C. R. Gerfen, R. Paletzki, N. Heintz, Neuron 80, 1368–1383 (2013).6. H. Hioki et al., J. Neurosci. 33, 544–555 (2013).7. H. Hama et al., Nat. Neurosci. 18, 1518–1529 (2015).

8. M. Ohkura et al., PLOS ONE 7, e51286 (2012).9. D. L. Ringach et al., Nat. Commun. 7, 12270 (2016).10. S. Kondo, T. Yoshida, K. Ohki, Nat. Commun. 7, 13210 (2016).11. S. Song, P. J. Sjöström, M. Reigl, S. Nelson, D. B. Chklovskii,

PLOS Biol. 3, e68 (2005).12. T. K. Berger, R. Perin, G. Silberberg, H. Markram, J. Physiol.

587, 5411–5425 (2009).13. A. Hayar, M. T. Shipley, M. Ennis, J. Neurosci. 25, 8197–8208

(2005).14. Y. Yoshimura, J. L. Dantzker, E. M. Callaway, Nature 433,

868–873 (2005).15. H. Kazama, R. I. Wilson, Nat. Neurosci. 12, 1136–1144

(2009).16. S. Lefort, C. Tomm, J. C. Floyd Sarria, C. C. Petersen, Neuron

61, 301–316 (2009).17. Y. Ikegaya et al., Cereb. Cortex 23, 293–304 (2013).18. L. Cossell et al., Nature 518, 399–403 (2015).19. M. C. Ashby, J. T. Isaac, Neuron 70, 510–521 (2011).20. M. Li et al., Brain Res. Bull. 81, 107–113 (2010).21. Y. C. Yu et al., Nature 486, 113–117 (2012).22. S.-B. Paik, D. L. Ringach, Nat. Neurosci. 14, 919–925 (2011).23. G. Ohtsuki et al., Neuron 75, 65–72 (2012).24. Y. Li et al., Nature 486, 118–121 (2012).25. L. A. Elias, D. D. Wang, A. R. Kriegstein, Nature 448, 901–907

(2007).26. R. Yuste, A. Peinado, L. C. Katz, Science 257, 665–669

(1992).27. A. Peinado, R. Yuste, L. C. Katz, Neuron 10, 103–114 (1993).28. R. Yuste, D. A. Nelson, W. W. Rubin, L. C. Katz, Neuron 14, 7–17

(1995).

ACKNOWLEDGMENTS

The AAV vector was a gift from A. Yamanaka (Nagoya University;CREST, Japan Science and Technology Agency) and is available

from Nagoya University under a material transfer agreement.We thank C. Yokoyama for thoughtful discussions andmanuscript editing, A. Miyawaki and H. Hama for valuablesupport with the anatomical experiments, S. Kondo and K. Ohkifor useful advice on in vivo microscopy, S. Tonegawa andH. Okamoto for helpful discussions on manuscript preparation,and H. Kazama and E. I. Moser for critical reading. We alsothank K. Kiso, N. Matsumoto, E. Ohshima, and M. Kishino fortechnical assistance and the RIKEN Brain Science Insitute–Olympus Collaboration Center for providing imaging equipment.This work was supported by research funds from RIKEN toT.H. and Grants-in-Aid for Scientific Research from MEXT(the Ministry of Education, Culture, Sports, Science andTechnology of Japan) to T.H. (Innovative Areas “MesoscopicNeurocircuitry,” 22115004), N.N. (16K14565), S.T. (24700344),and S.S. (25890023). Neuron coordinate data are availablein the supplementary materials. H.M., N.N., S.T., S.S., andT.Y performed the Ca2+ imaging. N.N. performed the gapjunction experiments. S.T. performed the synaptic connectivityexperiments. S.S. performed the anatomical experiments. T.Y.performed the visual response analyses. H.M., N.N., S.T.,S.S., T.Y., and T.H. analyzed the data and wrote the paper. T.H.conducted the research.

SUPPLEMENTARY MATERIALS

www.sciencemag.org/content/358/6363/610/suppl/DC1Materials and MethodsFigs. S1 to S7References (29–39)Data S1

19 January 2017; accepted 25 September 201710.1126/science.aam6125

Maruoka et al., Science 358, 610–615 (2017) 3 November 2017 6 of 6

RESEARCH | RESEARCH ARTICLEon M

arch 17, 2020

http://science.sciencemag.org/

Dow

nloaded from

Lattice system of functionally distinct cell types in the neocortexHisato Maruoka, Nao Nakagawa, Shun Tsuruno, Seiichiro Sakai, Taisuke Yoneda and Toshihiko Hosoya

DOI: 10.1126/science.aam6125 (6363), 610-615.358Science 

, this issue p. 610Sciencedevelopment is lineage-independent but guided by local electrical transmission.

specific gap junctions, suggesting that their−microcolumns developed from nonsister neurons coupled by cell type Microcolumns received common presynaptic inputs and showed synchronized activity in many cortical areas. These

cortical projection neurons, also form microcolumns that interdigitate with those of the subcerebral projection neurons. microcolumns make up a hexagonal lattice with a regular gridlike spacing. The other major layer 5 excitatory cell class,

examined large regions of mouse brain layer 5 and observed that thousands of these et al.microcolumns. Maruoka Subcerebral projection neurons, a major excitatory cell type in neocortical layer 5, form small cell clusters called

The fundamental organization of excitatory and inhibitory neurons in the neocortex is still poorly understood.The basic modules of the neocortex

ARTICLE TOOLS http://science.sciencemag.org/content/358/6363/610

MATERIALSSUPPLEMENTARY http://science.sciencemag.org/content/suppl/2017/11/02/358.6363.610.DC1

REFERENCES

http://science.sciencemag.org/content/358/6363/610#BIBLThis article cites 38 articles, 5 of which you can access for free

PERMISSIONS http://www.sciencemag.org/help/reprints-and-permissions

Terms of ServiceUse of this article is subject to the

is a registered trademark of AAAS.ScienceScience, 1200 New York Avenue NW, Washington, DC 20005. The title (print ISSN 0036-8075; online ISSN 1095-9203) is published by the American Association for the Advancement ofScience

Science. No claim to original U.S. Government WorksCopyright © 2017 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of

on March 17, 2020

http://science.sciencem

ag.org/D

ownloaded from